Sales of the NVIDIA GeForce RTX 4080 graphics card will start tomorrow, and today many specialized Internet sources published reviews of the new product. Here we present the most important excerpts from these materials, and in short: the GeForce RTX 4080 has become much faster than not only its predecessor, but also the previous generation flagships. However, the greatly increased price clouds the overall impression.
First, let’s recall that the GeForce RTX 4080 is based on a 4nm AD103 GPU with 9728 CUDA cores, 304 Tensor cores and 76 RT cores. The GPU clock of the reference version of the RTX 4080 is between 2205 and 2508 MHz. The graphics card is equipped with 16 GB of GDDR6X memory at 2800 MHz and a bandwidth of up to 22.4 Gb/s per pin – a total bandwidth of 716.8 GB/s. The suggested price is $1199. Note that NVIDIA initially announced a cheaper version of the GeForce RTX 4080 with 12GB of memory and a less powerful GPU, but later canceled it.
Reviewers unanimously noted the high performance of the GeForce RTX 4080. According to Tom’s Hardware, the new product proved to be 51% faster than its predecessor in games with 4K resolution compared to the GeForce RTX 3080. The new RTX 4080 also outperformed the flagship of the previous generation GeForce RTX 3090 in games by 21% Ti and 22% behind the current flagship RTX 4090.
Compared to AMD graphics cards, the GeForce RTX 4080 was 29% faster than the Radeon RX 6950 XT in Tom’s Hardware gaming tests. So if AMD’s statements about the superiority of the Radeon RX 7900 XTX by 50-70% over its predecessor prove true, then the GeForce RTX 4080 could face serious difficulties next month.
The GeForce RTX 4080 performs very well in ray tracing games, which is not surprising. It’s 26% faster than the RTX 3090 Ti, 72% faster than the RTX 3080, and more than twice as fast as the RX 6950 XT. And enabling DLSS gives you even faster speeds – allowing the new product to overtake the RTX 4090 if you don’t turn on DLSS on the last one.
At a lower resolution, the gap between the GeForce RTX 4080 and the flagship RTX 4090 has narrowed, as has the gap between the new product and previous-generation NVIDIA and AMD cards. Users of monitors with Full HD resolution will notice the slightest difference. But in general it would probably be strange to use a graphics card sharpened for 4K with a resolution four times lower.
In terms of power consumption, the GeForce RTX 4080 performs better here than the flagship GeForce RTX 4090. The reference version of the GeForce RTX 4080 Founders Edition is specified as power consumption of 320 W, but the card only consumes around 300 W under load, both in games and in stress tests. This is significantly less than the GeForce RTX 3080 in the same scenarios. According to TechPowerUp, the peak consumption is 371 watts.
Note that Tom’s Hardware managed to stably run the GPU of the GeForce RTX 4080 card at 3039.4 MHz, and at the same time the consumption was 307.9 watts. However, this is the case in games, while in FurMark the stable frequency during overclocking was 2621.5 MHz, while consumption increased to 332.1 watts. As noted by the reviewers, the novelty is highly efficient and offers high FPS per watt of energy consumed.
Overall, reviewers praised the GeForce RTX 4080’s capabilities and called it a good choice for a gaming PC with a 4K monitor. It proved better than the flagship in terms of performance per watt and per dollar. At the same time, the price of the GeForce RTX 4080 is still very high, especially compared to its predecessors – the previous GeForce RTX 3080 was rated by NVIDIA at $699, while the new product will have to pay at least $1199. In this context, it is interesting to compare the new product with the Radeon RX 7900 XTX, which AMD positions as a direct competitor.